The theory and outcome measure Failure-to-Rescue (FTR) was developed to assess hospital quality of care for surgical patients. This proposal seeks to extend FTR to the analysis of medical conditions. FTR measures the ability of caregivers to manage a patient who becomes complicated and keep them from dying. The measure depends on the ability of the analyst to differentiate complications from comorbidities and on the completeness of the available data in detecting complications. Due to difficulties in distinguishing complications from comorbidities, FTR has been applied almost exclusively to surgical cases. Since the metric FTR was first introduced by Silber in 1992, there have been many applications of the metric for quality assessment of surgical care, and it has been a useful metric for exploring the importance of hospital characteristics, nursing characteristics, and physician characteristics. FTR has been endorsed in multiple forms by the National Quality Forum. Unfortunately, the benefits of the FTR metric, both with respect to improving severity adjustment and gaining insight into why a hospital or provider may be performing poorly on mortality metrics, have never been successfully applied to patients with medical conditions. However, there has been a radical change in the data sets available to analysts since 1992 when FTR was first developed. Medicare claims data now provide a present on admission (POA) indicator for all diagnoses, which could greatly improve the ability of analysts to distinguish complications from comorbidities when implementing a medical FTR analysis. Therefore, it is now possible to potentially develop and validate a medical FTR metric. Given the demand for improved quality metrics by both the public and health policy analysts alike, it would seem like an ideal time to extend the measure to medical conditions. This project has 4 aims: AIM 1: Using national Medicare data and state data from California, we will develop and validate a list of FTR complications for major medical conditions including CHF, AMI, and Pneumonia. AIM 2: Using results from AIM 1, develop a medical FTR metric for each medical condition and study the properties of the FTR metric. We will explore 2 types of FTR rates: The original Silber defined approach (FTR) using all complications and the AHRQ approach (A-FTR) using a subset of complications and deaths, and we will compare results for all subsequent hypotheses. AIM 3: Develop a severity adjustment model for medical FTR and A-FTR, and finally, AIM 4: Determine the reliability and validity of the medical FTR metric. In summary, this proposal will establish a new approach to examining quality of care for medical conditions based on Failure-to-Rescue. Our intent is to bring to the medical community the advantages of using a failure-to-rescue analysis in a parallel manner to surgical patients.